C 3 R 1 C 1 C 2. Si- Trigger R 2 R 4. diamond. = 470 pf/1000v C 3 = 4.7 nf/1000v. Collection Distance [µm] Time [year] collected charge[e] - PDF

Description
EUROPEAN LABORATORY FOR PARTICLE PHYSICS CERN/EP 98-8 April 2, 1998 Development of CVD Diamond Radiation Detectors The RD42 Collaboration W. Adam 1, C. Bauer 2, E. Berdermann 3, F. Bogani 4, E. Borchi

Please download to get full document.

View again

of 10
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Information
Category:

Math & Engineering

Publish on:

Views: 26 | Pages: 10

Extension: PDF | Download: 0

Share
Transcript
EUROPEAN LABORATORY FOR PARTICLE PHYSICS CERN/EP 98-8 April 2, 1998 Development of CVD Diamond Radiation Detectors The RD42 Collaboration W. Adam 1, C. Bauer 2, E. Berdermann 3, F. Bogani 4, E. Borchi 5, M. Bruzzi 5, C. Colledani 6, J. Conway 7, W. Dabrowski 8,P. Delpierre 9, A. Deneuville 1, W. Dulinski 6,B.van Eijk 11,A.Fallou 9, D. Fish 7,F.Foulon 12,M.Friedl 1, K.K. Gan 13, E. Gheeraert 1, E. Grigoriev 2, G. Hallewell 9, R. Hall-Wilton 14,S.Han 15, F. Hartjes 11, J. Hrubec 1, D. Husson 6, H. Kagan 13, D. Kania 16, J. Kaplon 8, R. Kass 13, K.T. Knope 2, M. Krammer 1,P.F. Manfredi 17, D. Meier 8;y, M. Mishina 18, F. LeNormand 6, L.S. Pan 19,H.Pernegger 1,M.Pernicka 1, S. Pirollo 5,V.Re 17, J.L. Riester 6,S.Roe 8, D. Ro 14, A. Rudge 8,S.Schnetzer 7, S. Sciortino 5, V. Speziali 17, H. Stelzer 3, R. Stone 7, R.J. Tapper 14, R. Tesarek 7, G.B. Thomson 7,M.Trawick 13,W.Trischuk 2,R.Turchetta 6, A.M. Walsh 7,R.Wedenig 8, P. Weilhammer 8, H. Ziock 15, M. Zoeller 13 1 Institut fur Hochenergiephysik der Osterr. Akademie d. Wissenschaften, A-15 Vienna, Austria 2 MPI fur Kernphysik, D-6929 Heidelberg, Germany 3 GSI, Darmstadt, Germany 4 LENS, Florence, Italy 5 University of Florence, Florence, Italy 6 LEPSI, CRN, Strasbourg 6737, France 7 Rutgers University, Piscataway, NJ 8855, U.S.A. 8 CERN, CH-1211, Geneva 23, Switzerland 9 CPPM, Marseille 13288, France 1 LEPES, Grenoble, France 11 NIKHEF, Amsterdam, Netherlands 12 Centre d'etudes de Saclay, Gif-Sur-Yvette, France 13 The Ohio State University, Columbus, OH 4321, U.S.A. 14 Bristol University, Bristol BS8 1TL, U.K. 15 Los Alamos National Laboratory, Research Division, Los Alamos, NM 87545, U.S.A. 16 Lawrence Livermore National Laboratory, Livermore, CA 9455, U.S.A. 17 Universita di Pavia, Dipartimento di Elettronica, 271 Pavia, Italy 18 FNAL, Batavia, U.S.A. 19 Sandia National Laboratory, Livermore, CA 9455, U.S.A. 2 University of Toronto, Toronto, ON M5S 1A7, Canada Abstract Diamond is a nearly ideal material for detecting ionizing radiation. Its outstanding radiation hardness, fast charge collection and low leakage current allow a diamond detector to be used in high radiation, high temperature and in aggressive chemical media. We have constructed charged particle detectors using high quality CVD diamond. Characterization of the diamond samples and various detectors are presented in terms of collection distance, d = E, the average distance electron-hole pairs move apart under the inuence of an electric eld, where is the sum of carrier mobilities, E is the applied electric eld, and is the mobility weighted carrier lifetime. Over the last two years the collection distance increased from 75 m toover 2 m. With this high quality CVD diamond a series of micro-strip and pixel particle detectors have been constructed. These devices were tested to determine their position resolution and signal to noise performance. Diamond detectors were exposed to large uences of pions, protons and neutrons to establish their radiation hardness properties. The results of these tests and their correlation with the characterization studies are presented. presented at 5 th Int. Symposium on Diamond Materials, Paris, (1997). y corresponding author D. Meier: 1 Introduction Solid state tracking devices have become one of the main stays of general purpose high energy physics detectors. Detectors in future high energy and nuclear collider experiments will be exposed to high radiation levels. At a distance of 1 cm from the beam axis, for instance, detectors are expected to receive a uence of 1: particles per cm 2 during 1 years of operation at the Large Hadron Collider (LHC) at the European Laboratory of Particle Physics (CERN). There are very few materials which can withstand this level of radiation. CVD diamond is a radiation resistant detector material which may be able to operate close to the interaction region at future experiments. The RD42 collaboration is working as a detector R&D project on the development of diamond detectors for particle tracking at LHC. We study CVD diamond with regard to their electrical material properties, their radiation hardness and tracking performance and participate in the development of fast and radiation hard readout electronics. We have built various diamond sensors and tracking devices in strip or pixel geometries which have been studied with readout electronics presently in use or under development for LHC. The CVD diamond substrates are manufactured in industry in size and geometry comparable to silicon detectors. Dierent from silicon detectors, they operate in high radiation environment at room temperature with negligible leakage current and negligible power dissipation. Noise due to leakage current is negligible in the insulating diamond detector. The capacitance of diamond is low due to relatively low electrical permeability resulting in a reduced noise. The low proton number of diamond causes relatively small multiple scattering of particles which is important for particle tracking and exact identiction of decay vertices. By working with CVD diamond manufacturers we have been able to improve the diamond substrate over the last few years. In the following we are reporting the work on diamond characterization, performance of a characteristic tracking device and studies on neutron, pion and proton irradiations. 2 Charge Collection Properties of Diamond Detectors We characterize CVD diamond samples in terms of their charge collection distance which is the distance electrons and holes move in average apart in an external electric eld inside the diamond. It is one of the most important characteristics for diamond-based particle detectors because of its relation to fundamental material properties. The distances d e, d h which electrons or holes travel in an electric eld E depend on their lifetime e, h and mobility e, h. d e = e e E and d h = h h E; (1) The charge collection distance d is the sum of these distances and can be expressed using the summed mobility = e + h and the mobility weighted lifetime of electrons and holes d = d e + d h E : (2) Above a certain eld strength the carrier velocities saturate and d is expected to stay constant. It was shown previously, that the charge collection distance varies as a function of depth within a sample and particularly increases from the nucleation to the growth side [1, 2]. We have expanded this observation by measuring the charge collection distance for high quality diamond after successive material removal from the nucleation side and observe a linear increase in the collection distance. In another study we exposed diamond samples to relatively small doses of radiation and observe an increasing charge collection distance. This \pumping eect has been observed previously [3]. In the pumped state traps are lled and release their charge during either heating or exposure to `cold' uorescent light. This release of charge is also observed by luminescence light and thermally stimulated current measurements. 2.1 Status of Charge Collection Distance Fig. 1 shows a basic diamond readout and how its charge signal is measured. The signal can be read on both sides of the detector. The charge signal is either generated by electrons from a source or by high energetic particles from an accelerator. Charged particles traversing the 9 Sr R 1 C 3 U Si- Trigger mm 2 d 5 mm C 1 C 2 electron diamond R 2 C 1 C 2 R 4 VA2 readout = 47 pf/1v C 3 = 4.7 nf/1v = 47 pf/1v R 1 R 2 =1 M Ω d = 1 µ m... 2 mm U=..1 kv Figure 1: Principle and setup for charge collection measurement in a diamond detector. In the lab charged particles from radioactive sources are used to measure charge collection of the sample The diamond and readout electronics in this conguration is also used for particle tracking in high energetic beams. collected charge[e] electric field [V/µm] Figure 2: Mean collected charge and charge collection distance on CVD diamond from recent growth as a function of the applied electric eld. The charge collection distance on this sample reaches d = 25 m. 5 diamond detector deposit energy and generate electron hole pairs along their path. The charge Q col which can be collected at the electrodes after one charged particle traversed the diamond characterizes the performance of the detector. A minimum ionizing particle mip, e.g. an electron with a kinetic energy of 3 MeV, deposits 245 kev in D = 5 m thick diamond. The energy which is necessary to ex cite one eh-pair in diamond is 13.6 ev. One 3 obtains the number hq mip;gen i =D =36e=m 25 of generated electrons or holes [4]. Electron 2 hole pairs separate in the applied electric eld 15 and travel towards the electrodes where they 1 induce a charge Q col. The mean measured 5 charge hq col i is related to the charge collection distance d of the diamond bulk and in an Time [year] approximation one nds [4] Figure 3: Historical evolution of charge collection in CVD diamond measured with the de- hq col i d hq mip;gen i = hq coli =D 36 e=m ; (3) scribed method [Fig. 1]. Collection Distance [µm] which relates the charge collection distance to the collected charge at the electrodes. Fig.2 shows a measurement of the charge collection distance as a function of the applied electric eld. The charge collection distance on this samples reaches d = 25 m at 1 V/m. The samples has a thickness of 198 m. The charge collection saturates between.8 to 1 V/m. Using an electron mobility of 18 m 2 /V/ns and a hole mobility of 12 m 2 /V/ns we estimate for this sample a saturation of carrier velocities at 25 m/ns and a carrier lifetime of about 1 ns. Fig. 3 shows the increase of charge collection distance on CVD diamond over the past 7 years and the charge collection of 25 m on available CVD diamond today (1997). 2.2 Material Removal Study CVD diamond typically grows in a polycrystalline columnar structure along the growth direction. The substrate side begins with small grains ( 1 m) which grow with material thickness. As the material grows it develops the texture of the fastest growing crystal orientation. It has previously been shown [1, 2] that the electrical properties of CVD diamond vary with the thickness of the material: the carrier lifetimes are small on the substrate side and large on the growth side. As a result, the raw diamond material can be \improved by removing material from the substrate side. This procedure has been used to increase the signal size by 4% over the as-grown sample. Fig. 4 shows the improvement of the quality of diamond measured by the increase in signal size as a function of the fraction of material removed. The measurement can normalized charge collection distance [ ] removed material [%] Figure 4: Measured charge collection distance as a function of the material removed from the substrate side. The sample was lapped three times by about 6 m at each step. The charge collection increases after each step which isex- plained by the linear increase of charge collection from the substrate side in the diamond. be explained by assuming a linearly increasing charge collection distance d(x) = cxalong the direction of growth x with a slope c =2d original =D original and d() = on the substrate side of the sample. The charge collection distance d 1 after material removal D 1 from the substrate side leaving a sample of thickness D = D original D 1 is then given by d 1 = R Doriginal D 1 R Doriginal D 1 d(x)dx dx = d original 1+ D 1 D original! ; d D original (4) The increase of charge collection distance predicted by Eq. 4 is shown in Fig. 4 in good agreement with the measurement. 2.3 Pumping with Charged Particles It is known that the charge collection distance at constant voltage increases under illumination with or radiation [3, 5]. On unirradiated samples an increase by a factor of 1.6 to 2. below 1 Gy absorbed dose is observed. Above 1 Gy charge collection distance stays constant [Fig. 5]. The increase at relatively low dose is called pumping. It is attributed to trap lling which then allows charge carriers to travel longer distances before being traped which is equivalent to an increase in lifetime [Eq. 2]. A sample can be depumped by exposure to uorescent light and charge collection distance returns to its original value [6]. Depumping is explained by releasing lled traps. The eects of pumping and depumping are reproducible. After exposure to 9 Sr of dierent dose we observe luminescence light and excess current during heating [Fig. 6,7]. The intensity of the luminescence light increases linearly with dose and saturates. Exposure to light decreases luminescence and thermally stimulated current as well as charge collection distance. ccd d c [µm] time [min] Figure 5: Pumping on a diamond with - radiation from a 9 Sr source. The charge collection distance increases and saturates under illumination with charged particles. Thermoluminescence on Diamond, Dose Dependence current[µa] time [min] Temperature [K] TL Intensity [arbitrary] Temperature [ o C] Figure 6: Temperature stimulated current as a function of temperature after exposure to - radiation from a 9 Sr source. The current increases from a few pico-ampere at room temperature and 1 V to several 1 A during heating. Once heated, charge carriers are ushed and the current peak is missing. Figure 7: Thermo-luminescence signal as a function of temperature at dierent doses of - radiation from a 9 Sr source. The dotted graph shows the thermo-luminescence after one hour pumping with 9 Sr followed by an exposure to a uorescent light lamp. 3 Tracker Studies The rst diamond particle tracker was tested in 1993 at CERN [7]. It had 1 m readout pitch and a charge collection distance of 5 m. In these tests a 1 GeV/c pion beam ( ) passes through a telescope consisting of eight planes silicon micro-strip detectors. The silicon telescope predicts the pion track within a few micro-meter resolution. The diamond detectors under study are in between the silicon planes. Fig. 8 shows a metalization pattern on a recent strip tracker. The strips have a pitch of 5 m. They form ohmic contacts to the diamond. Other patterns like pixel or pad structures were prepared and tested as well. The charge signal on strips is read out using VA2 VLSI chips [8]. The VA2 contains 128 channels of low noise charge integrating preampliers followed by a signal shaper. The shaped signal has a peaking time of about 2 s and in praxis an equivalent noise charge of 8 e+11 e=pf. Other readout congurations which suit requirements at the LHC like shorter peaking time and radiation hardness are being studied as well [9]. The charge collection distance on the diamond tracker described here was measured with a 9 Sr source in the lab [Fig. 9] and reaches about 21 m at an electric eld of 1 V/m which corresponds to 76 e collected charge. Fig. 1 shows the distribution of collected charge on this diamond as a strip detector in a 1 GeV/c pion beam. The charge signal is nearly Landau distributed and has a mean value of 82 e depending on the number of strips adjacent to the predicted intersection. The mean collected charge of 82 e corresponds to a charge collection distance of 23 m which agrees within the measurement error with the 9 Sr result. The noise on a strip ideally depends only on the leakage current and strip capacitance. In diamond the contribution due to leakage current is negligible. The capacitance is of a few pico-farad depending on the strip geometry. We measured a noise of about 15 e which gives a mean signal-to-noise ratio for this tracker of 55-to-1. The distribution of the dierence between the track prediction perpendicular to the strips and the actual measured position by the diamond detector is nearly gaussian and gives a resolution of = 16:5 m [Fig. 11] close to the digital resolution expected for a strip detector with 5 m readout pitch p ( digital =5m= 12 14:3 m). Figure 8: Photo of a diamond strip detector. The strips have a width of 25 m and pitch of 5 m. The magnication right at the bottom shows bond pads from were bond wires connect to the readout chip. Collection Distance (µm) Electric Field (V/µm) Mean Charge (e) Figure 9: Charge collection distance measured with a 9 Sr source and dot electrodes, before it was metalized with strips and prepared as a tracker. 4 Irradiation Studies Radiation hardness is required for particle detectors in e.g. future particle physics experiments at the LHC/CERN or Tevatron/FNAL. In particular solid state tracking detectors have to resist high particle uences keeping the signal-to-noise ratio as high as possible after irradiation. Solid state devices are damaged under particle irradiation. Damage in solid state detectors causes on one hand an increase in leakage current and therefore an increase in noise; on the other hand a reduction in the amount of collected charge, which leads to a smaller signal. The signalto-noise ratio in damaged detectors decreases. The potential advantage of diamond is the much larger dose that is apparently necessary to cause the same damage. events [ ] charge signal [1 3 e] events [ ] u hit -u track [µm] Figure 1: Pulse height distribution of the signal charge on strips next to the predicted track. Figure 11: Distribution of the dierence between the measured hit in the diamond detector and the predicted track intersection. 4.1 Neutron Irradiation CVD diamond samples and silicon diodes were irradiated with neutrons in four exposures during the last two years using the ISIS facility at the Rutherford Appleton Laboratory, England. Protons from an accelerator interact in a graphite block and generate neutrons ranging from 1 kev to several 1 MeV peaking at 1 MeV [1]. The mean n-ux above 1 kev was about (1:7 :6) 1 8 n=cm 2 =s. The samples were kept at room temperature during irradiation. The samples were biased at 1 V and we could measure their leakage current during irradiation [Fig. 12]. In diamond we observe a neutron or -background induced current 3 pa which is correlated to the neutron ux. The neutron ux decreases as soon as the accelerator stops delivering protons to the spallation target which results in a prompt decreasing current in the diamond to a few pico-ampere. On samples which were irradiated for the rst time the induced current decreases exponentially and saturates on a level of several 1 pa. The leakage current in silicon diodes on the other side increases continuously from a few micro-ampere to several 1 A. During spill breaks when no neutrons are present one observes a slightly decreasing current in silicon (annealing). The charge signal distribution on a diamond samples before and after n=cm 2 is shown in Fig. 13. The most probable value of the signal charge ordinate and units at graph bias voltage [V] on Silicon Diode current [µa] in Silicon Diode current [ 1 pa] in Diamond U7 at 1 V time [h] Figure 12: Induced current in diamond and in a silicon diode during neutron irradiation. The current in diamond shows a prompt response to the presence of neutrons. The induced current goes back to a few pico-ampere during spill breaks when no neutrons are present. The current in Silicon increase continuously and anneals during spill breaks. does not change. The mean of the distribution after irradiation may have changed slightly but is within the measurement error compared to before irradiation. The charge collection distance normalized to the pumped value before irradiation is shown in Fig. 14. events/bin [ ] normalized charge collection distance [ ] signal charge [e] (ISIS spallation neutrons) fluence [1 15 cm -2 ] Figure 13: Charge signal distribution on a diamond detector sample before and after two n- irradiations. The sample acquired a n-uence of n=cm 2. Figure 14: Charge collection distance in CVD diamond detector samples as a function of neutron uence up to n=cm 2. The neutrons come from the ISIS spallation source and have a kinetic energy spectrum peaking at 1 MeV. 4.2 Proton Irradiation CVD diamond sampl
Related Search
Similar documents
View more...
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks